DC Field | Value | Language |
---|---|---|
dc.contributor.author | Ku, Hyun Jin | ko |
dc.contributor.author | Ha, Hyunho | ko |
dc.contributor.author | Lee, Joo Ho | ko |
dc.contributor.author | Kang, Dahyun | ko |
dc.contributor.author | Tompkin, James | ko |
dc.contributor.author | Kim, Min Hyuk | ko |
dc.date.accessioned | 2022-11-11T00:00:24Z | - |
dc.date.available | 2022-11-11T00:00:24Z | - |
dc.date.created | 2022-09-07 | - |
dc.date.created | 2022-09-07 | - |
dc.date.created | 2022-09-07 | - |
dc.date.created | 2022-09-07 | - |
dc.date.issued | 2022-08-03 | - |
dc.identifier.citation | International Conference on Computational Photography, ICCP 2022 | - |
dc.identifier.uri | http://hdl.handle.net/10203/299507 | - |
dc.description.abstract | Reconstructing 3D objects in natural environments requires solving the ill-posed problem of geometry, spatially-varying material, and lighting estimation. As such, many approaches impractically constrain to a dark environment, use controlled lighting rigs, or use few handheld captures but suffer reduced quality. We develop a method that uses just two smartphone exposures captured in ambient lighting to reconstruct appearance more accurately and practically than baseline methods. Our insight is that we can use a flash/no-flash RGB-D pair to pose an inverse rendering problem using point lighting. This allows efficient differentiable rendering to optimize depth and normals from a good initialization and so also the simultaneous optimization of diffuse environment illumination and SVBRDF material. We find that this reduces diffuse albedo error by 25%, specular error by 46%, and normal error by 30% against singleand paired-image baselines that use learning-based techniques. Given that our approach is practical for everyday solid objects, we enable photorealistic relighting for mobile photography and easier content creation for augmented reality. | - |
dc.language | English | - |
dc.publisher | IEEE | - |
dc.title | Differentiable Appearance Acquisition from a Flash/No-flash RGB-D Pair | - |
dc.type | Conference | - |
dc.identifier.wosid | 000869734300001 | - |
dc.type.rims | CONF | - |
dc.citation.publicationname | International Conference on Computational Photography, ICCP 2022 | - |
dc.identifier.conferencecountry | US | - |
dc.identifier.conferencelocation | Caltech, Pasadena | - |
dc.contributor.localauthor | Kim, Min Hyuk | - |
dc.contributor.nonIdAuthor | Lee, Joo Ho | - |
dc.contributor.nonIdAuthor | Tompkin, James | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.